REVIEWS— INTRODUCTION TO CRYPTOGAMIC BOTANY. 343 



to ttem. The author recognizes only three grand divisions — Exogens, 

 Endogens, and Cryptogams. He says, in reference to Lindley's 

 system : " I cannot consider Dictyogens (much less Rhizogens and 

 Gymnogens) as a class of the same importance with Endogens and 

 Exogens. They are so clearly endogenous, notwithstanding the 

 peculiarities of the venation, or much more of the structure of the 

 stem, that unless every anomaly is to be considered as overthrowing 

 a natural division, we miist either be content to leave them in com- 

 pany with their allies, or give up the attempt of natural arrangement 

 altogether." Probably the advocates of the system referred to, do 

 not hold it to be necessary that all classes should be accounted of 

 equal importance ; their view is that smaller transition groups are 

 better separated as classes than only set apart as sub-classes of the 

 larger class which they most nearly approach. Eew for instance 

 would be satisfied with considering Grymnogens as no more separated 

 from other Exogens than as any one alliance, or even than as one 

 great section is separated from another. Those who do not admit 

 the class, divide exogens into sub-classes of very unequal extent, 

 Angiospermse and Grymnospermae. It can hardly be denied that 

 this distinction is real and important. The question is respecting 

 the best mode of expressing it, and we still incline to prefer Dr. 

 Lindley's plan of increasing the number of classes, though some of 

 them be obviously transition groups of smaller extent and less dis- 

 tinctly marked characters than the others. To us Dictyogens seem 

 a good deal more doubtful than Grymnogens, but we like the idea of 

 these transition classes, and accept for the present an arrangement 

 which includes them. Grrant that Grymnogens and Ehizogens have 

 a nearer relation to Exogens, Dictyogens to Endogens, and that 

 Acrogens and Thallogens may be connected together as Cryptogams, 

 yet if these divisions must be recognized at all as something more 

 than orders, the simplest way is to adopt them as classes, but without 

 holding the principle that all classes are of equal value any more in 

 the nature of their characters than in their extent. Having defined 

 cryptogams by the joint consideration of their (generally) cellular 

 substance, their growth by superficial development, the absence of 

 organs strictly corresponding with stamens and carpels, though 

 there are sexual organs — the general substitution of bodies resem- 

 bling spermatozoa for pollen grains, and the absence of a true embryo — 

 he proceeds to justify himself in retaining the familiar name Crypto- 

 gams in preference to several which have been proposed, in which 



